BREIN prediction feature

There is a new BREIN tab called ‘Prediction’. It predicts the performance of a final product based on the current state of the production.

The mechanism behind is the following: After the final traverse analysis the values of the fit parameters are passed to a CODE configuration which produces the prediction graph. Typically the numbers are layer thicknesses – this way the predicting CODE program knows the current layer stack and can compute the performance of any glazing product that includes the actual coating. So the operators can take a look at the final product including gas fillings and second or third panes.

The prediction is done for a single traverse position. It is triggered by the arrival of new results for this position only.

More flexible layer stack view

Introduced with object generation 3.97

Visualizing a layer stack in views is now more flexible. Instead of representing each layer as a rectangle of constant height (i.e. the same height for all layers) the graph can now better visualize the proportions by drawing the rectangles with heights proportional to the thickness of the layer.

The graphs below show the difference. Up to now layer stack views looked like this:


Now you can have this:


This kind of drawing makes it much easier to compare layer stacks:


The new kind of drawing is activated by a click on the layer stack in the view. You will be asked for the thickness that the height of the base rectangle should represent. The value must be specified in micrometers – if you set a value to 0.075 the base rectangle shows 75 nm of the stack.

Since layer thickness values in stacks with thin films and (incoherent) thick layers are very different, it is not possible to see thick layers and thin films in one graph. Thin films with thicknesses 100000 times smaller than thick layers will simply not be visible (like in real life). If you still want to show the thin film part of the stack with the new feature you have to use 2 layer stacks: one for the thin film coating alone and the second one for the complete layer stack. In this case you have to insert the coating as ‘included layer stack’ in the second one.

SPRAY surface textures

SPRAY is a powerful tool to optimize the performance of solar cells. In order to simplify the modeling of crystaline silicon PV modules we have added an easy way to define macroscopic surface texture. This new feature can be used for both the covering glass and the Si wafer surface.

A PV module consists of a silicon wafer with top and backside contacts:

wafer spray

The wafer is embedded in EVA:

wafer spray packed

The EVA is covered by glass which gives mechanical strength and protection for the next 30 years. In order to maximize the generation of electric power one may introduce an

  • anti-reflection (AR) layer between silicon and EVA
  • AR layer between glass and air
  • surface texture of the silicon wafer
  • surface texture of the cover glass
  • diffusely reflecting white backside
  • low-absorptive cover glass

In order to find out how much the PV output is increased by these improvements one can perform realistic ray-tracing computations with SPRAY. This can save a lot of experimental work and costs. SPRAY performs 3D spectral ray-tracing and records where how much radiation is absorbed. It can, for example, output the spectrum of absorbed light in the wafer

wafer spray result1


as well as the spatial distribution of absorbed light:

wafer spray absorption depth profile

The graph below shows a comparison of the absorbed fraction with and without a simple AR coating (a single SiN layer) between silicon and EVA:

wafer spray simple SiN AR


The modeling of macroscopic surface textures is now much easier using the new object type ‘Periodic surface texture’. This is a rectangle with a regular surface pattern in the x- and y-direction. You can set the periodicities in both directions and select a pattern type. Depending on the type of texture you can set additional parameters in the object dialog:



The available textures are the following:

Sine profile



Inverted pyramids
Inverted cones
Half spheres
Inverted half spheres

Improved start behavior

If BREIN was started after a long break during which a large number of results have been generated, the software was not responsive for a long period of time (because it was reading many result files). You can now limit the number of result files to be processed during one timer event (which happens every second). A number of 20 is reasonable and has been set as default value. You can change this choice in the Settings submenu.

Old filename menu items

Object generation 3.96

The management of the menu items that show the last accessed files has been improved. Since the list of files is now stored in a different way the menu items are all empty when you change to the new version. During your work the items will fill up again.

Use sheet resistance values

You can now use measured sheet resistance values for analysis of traverse data. This can significantly stabilize the fitting of silver based glass coatings, in particular if you fit both silver thickness and quality (using the Drude model damping constant).

In order to use sheet resistance values, you need to update VAAT_spectrum_scanner.exe, bright_eye_traverse.exe and the OLE server code.exe. All program files must be generated on August 29, 2014, or later.

Please read this as well: CODE link

Archive options

While BREIN is running, it produces a lot of files in the archive and the database folder. These directories are subfolders of the BREIN program directory (up to now). In order to avoid that this directory blows up over time you can use the following options to store all permanent data outside the program folder. Within the BREIN program folder only temporary files are used – so it should not increase in size any more. Make sure that the Cleaner program is running, and connect BREIN to an external SQL database if you can.

The archive can now be located anywhere in your file system – it need not be a subdirectory of the BREIN program folder.

You can switch off writing result files to the archive.

You can switch off writing scan files to the archive.

The database history files can be stored to the archive as well, instead of being stored in the product folder within the database.